Oracle DM Process
Data mining is a process that analyzes data from different viewpoints and produces information sets organized in a concise manner. Applications then use this information for a variety of purposes. For example, the information could demonstrate methods to increase productivity and revenue or reduce costs and waste. Data mining finds connections or patterns between different fields in databases. Oracle uses a process to harvest the data that protects the integrity of the original information.
Other People Are Reading
-
What Is Data Mining
-
Also known as knowledge discovery in data, the goal of data mining is to discover patterns, predict likely outcomes and use large databases of information for action. The process accomplishes these goals by using mathematical algorithms to dissect or break down information. Data mining then evaluates the probabilities that exist for future events based on the information.
Problem Definition
-
The Oracle DM process begins by attempting to understand the objectives and requirements of a given project and then developing an implementation plan. In other words, the Oracle DM process asks questions about the problems the organization faces and creates a model that produces possible outcomes. Before organizations can use this model, however, they must gather the information about likely relationships.
-
Preparation and Gathering
-
The preparation of the data gathering requires an understanding of what is involved in the data collection and exploration. This represents the part of the process that scrutinizes the data and determines how it addresses the problem. At this stage of the preparation, the organization adds and removes data based on the quality of the information. The process also uses different attributes to ask and answer questions that give the desired results.
Model Building, Evaluation and Deployment
-
The modeling phase uses small data samples to create a preliminary model. The process then tests this preliminary model before creating the final model. Since the final model typically contains a significant amount of data, the organization should verify the preliminary model before executing the final model. The organization should specifically evaluate the preliminary model and determine if it meets the end goals for the project. During deployment, the organization assesses information derived from the data for insight and actionable information. The deployment is typically used to provide reports and implement new methods of operation.
-
